Tuesday, 6 January 2015

Smart TVs will never succeed - What would be better?

I have had this same discussion few times in various threads in Google+ in the past years. Now, as we see new connected TV & Android TV products popping up from the CES I think it is a good time for me to explain why I don't believe any of these products will ever be popular.

I believe that smart TVs are a bad idea but connected TV is a good idea.

The reason I say this is that I see the concept of a smart TV being flawed. I don't believe there's a way to make the TV act as the center of the living room in the era of internet content.

Let's take a look how we used to use our TVs in the past (and some still do).

There is a two dimensional list of channels which you navigate either by typing in the the channel number or navigating channel up / channel down until you find something you find interesting (or at least become bored of changing the channel). I'd argue that most user use the up/down buttons more than typing the numbers (although I have no data to back this up and might be completely wrong).

Buttons you actually use on your remote control are few and far between.

What are the rest of the buttons on the remote that still look like they've never been touched when the text on your volume buttons is long gone?

Manufacturers adding new and "improved" features to our TV is not a new thing. We've seen teletext, EPGs, DVR-functions, etc. Some of them have been used and some of them not so much.

Why didn't we use them? Why did we often chose to try to find information form a newspaper rather than reading the teletext page?

Because these features were inconvenient and often annoying to use.

Remote control is an indirect and high latency control mechanism. We adopted it as our primary controller for changing channels because it was adequate for the two dimensional content hierarchy we had. However I would go so far as arguing that if the same mechanism would be introduced today it would be rejected as bad UI and we would demand something better. "They are asking us to memorise numbers to find BBC World?" "No way!".

More than two dimensional content

Internet content comes in many forms and many sources. How do you comfortably navigate between the content? How do you find the YouTube video you want? Is adding an additional dimension to the controller enough? Is D-Pad the solution?

No it isn't. D-Pad navigation is cumbersome, unpredictable and slow. Even platforms like the Apple TV what are fully designed to be operated with the simple D-Pad controllers infuriate users and if an alternative is available they'll take it.

Livingroom PC?

Our computers are OK for finding content and using it. Maybe we should make our TVs more like them?

Qwerty in livingroom?

There are people who like this idea. I for one used to have a living room PC. I had my own streaming server that I was able to use to send movies to my TV and that was great. I also used to love running Plex on my Google TV. But I'm a geek. I don't see many people wanting to do this. The ones who want to do it have tons and tons of options to get their streaming server setups running on a lot of different hardware and software but this is not what we want to force on the masses.

Shared Device

TVs are shared devices. Many people use the same device. In the past this wasn't an issue as the only thing that you personalised on your TV was the channel order and most families could probably reach a solution that everyone was happy with.

But what about now? Most people don't feel comfortable leaving their phones unlocked on a table when there's guests in the house (in some cases even when there isn't). Our phones are very personal devices and we don't like other people fiddling with them. But now we want to put these same platforms to our TVs?

Whose account we'll use to login to the Google Play Store? Which Netflix account we should have there? What about YouTube? Do I like it if other people mess up the recommendation engine of my YouTube account?

Who wants to log in with their account to the shared device?

Do we want to start logging into our TVs every time we turn them on? I don't know about you but to me it doesn't sound like a good idea..

But what about games?

For me to argue that playing games on a TV is stupid would be stupid. I'm not a console gamer but a lot of people are. The success of Playstation and the like is undeniable. This is also probably why Google guided developer attention to Gaming by bundling a game control their latest generation developer box of their Android TV they gave away to participants in the Google IO event.

I believe that Android is a viable platform for Gaming. A quick look to the Google Play gaming selection should convince anyone about that. But is an Android Gaming console a good idea?

There's plenty of people trying this. Maybe one of them will succeed but I remain sceptical. I think there's a better way to do all of this!

Google has already solved all this!

I'd argue that Chromecast is the best idea that has come out of Google since.. well probably ever. It's simple and exactly what I wanted to see. Your TV is an extension of a device that is better suited for interaction, your phone or tablet.

Even though at the start of the Chromecast era the device was a unreliable and sometimes required rebooting, reconnecting and black magic for it to work the potential was immediately obvious. No clumsy interaction with the TV but all the content available on the best screen of the house. Sold!

Then shortly after became another home run from Google. Android Wear. 

While Wear is technically very different from Chromecast the key idea of another screen extending another is obvious. Importantly, there is no Google Play Store on the Wear. Your Wear device is coupled to your phone and what you have on your phone is available on your wearable (as long as the app also supports Wear).

Google has also emphasised that apps extend functionality of the phone counterpart and not act as standalone apps on the wearable as it very rarely makes sense to operate an app on the clumsy interface of the wearable... Hmm... Where have we heard this before?

Dumb TV, Smart Content, Gaming

I want Google to pursue the idea of other screens extending other devices instead of making them standalone. I want this in my livingroom. I'm already using Chromecast on multiple screens but gaming and Chromecast doesn't really work. Screen mirroring is not good enough solution games need to run on a separate box. But that box does not need to be smart. It needs to be a dumb terminal without a UI and without a Store.

What if I had an Android box connected to my TV (or integrated) that has no standalone interface. Exactly as the Chromecast. No user is ever directly logged into the box. No data is stored there. No direct control even possible. Or at least that is how it would look lik to the users. Technically you would probably have data stored on the TV box. But it's all about mental models and appearances.

The user experience of playing a game on my TV would be the following:

  • I use my phone or tablet to go to the Play Store and find the game I want to play. I press install. If the game has a TV compatible component it will then automatically be installed on my TV box under my account. Whenever I play the game on my portables the Google Games cloud save will store my game progress etc. 
  • I come home and want to play the game on a larger screen. I open the game on my phone and see the familiar Cast icon I've already got used to when using the Chromecast for videos. I press the icon.
  • The game launches on my TV box. It will not run from my phone but from the box. I will be logged in to my account automatically and my game progress is restored.
  • My phone goes into game controller mode showing me the best configuration of on-screen buttons to play the game I've started.
  • I pickup my gaming controller and press the "on" button. The TV box recognises my controller, configures it to my settings (as it knows who is using it) and my phone can now turn off the gaming controller mode.
  • I play the game on my TV. All progress is saved to the Google Play cloud save.
  • Once I've had enough I pull down the notification shade on my phone that shows me that I'm casting to the TV. I press the close button. The game closes on the TV and I see the familiar Chromecast backdrop. The TV is now available to others to use.
Technically, I believe Google could already do that. In fact their Android TV boxes can already run the games very nicely and controller support is there. We now just need to drop the Android TV UI and Play Store from these devices.

At no point I have to suffer the unbearable controllers we have for TVs. I don't need to login. I don't need to use search on D-Pad to find the games on Play Store or from my installed games list. I want this.

But, other apps can do the same

True. What would prevent other apps from appearing to the TV the same way games do? Nothing really. Other than very few apps really make any sense to be on TV as standalone. We have the same situation on Android Wear. We can create standalone launchers, calculators, whatever for Wear but they have not became popular as they don't make sense.

Beyond TV

I believe that Google has created a platform that has limitless possibilities. We need to stop thinking about devices the way we have used to. Apps need to start living in many screens. You need to be able to use the best screen available to you anytime.

What if I want to write a blogpost on my Android device? I can do it now on my Nexus 9 and that keyboard but what if I want to use an external screen for it. I can plug in my slimport adapter, external BT keyboard and a mouse and start writing.

Very nice, but it could be better.

Why can't I have a cheap Android box connected to my desktop and a BT keyboard next to it? I just open my Blogger app on my phone, cast it to my desktop and start typing away. No need for controllers etc.

I want us to start decoupling apps from screens.

Saturday, 27 December 2014

My Thoughts on the Nexus 9 Tablet

I've now been using the Nexus 9 for a month or so. Few weeks ago I also finally managed to get the keyboard case delivered to me. I thought now would be a good time to share some thoughts about the device.

Let's get one thing out of the way first. No, I'd not recommend you to buy it.. unless you're in a certain special group of users. Let me explain.

The Nexus 9 is way too expensive for what it is. It's an Android tablet, one of the best ones, but it's way too expensive compared to alternatives. There are many things I don't like about the thing but the aspect ratio especially.

4:3 aspect ratio

This aspect ratio doesn't make any sense in most use cases. All (well almost all) videos are shown in a letter box and the screen space for a video is nearly identical to the much smaller and cheaper Nexus 7. 

Nexus 9's screen aspect ratio really isn't very good for video or reading. On nexus 7 the video size is almost same as on the N9.
iPad mini included for comparison. Personally, I think that iPad mini's size doesn't make any sense at all.
For reading and web the wide-screen format is superior as well. Holding the N7 in one hand is very comfortable while the N9 is not. Line length on the N9 is too long making reading experience uncomfortable. 

BUT -  The Keyboard

There's a but in everything and there's a but in the aspect ratio, price and holdability of the N9 as well.

The keyboard dock. It literally transforms this device (yes, I am typing this review on my Nexus 9). All the sudden the 4:3 aspect ratio makes sense. Google Docs feels like made for this device with this keyboard. To me Nexus 9 combined with the keyboard cover is the perfect travel device.

The higher angle of the dock makes the tablet very comfortable when using on a desk.

With the lower angle configuration the dock allows comfortable on-lap typing.

The small size of the tablet makes it easy to write on a plane even if you're stuck with the middle seat.

It's still not cheap. You'll have to pay more than 500 euros for the tablet & keyboard combo. But it depends where you compare it to. A small Macbook Air costs more and is less useful if you already own a developer device (a MBP in my case). A MBA is also much more cumbersome in small space like airplane table. I'm also much more comfortable with Android than any desktop OS on my devices. It is better thought out and more beautiful than any of the alternatives.

This is where I'll change my to-buy recommendation to yes. If you're looking for a device that lets you write on the go I doubt you'll find a better device than the Nexus 9 in the price range.

The keyboard is very comfortable to type with. It is smaller than you laptop / desktop equivalent but it's not too small. The 2 different positions allow you to use the keyboard either on lap or on a table. 

The aspect ratio defines the device's purpose more than you would think. The device is great for writing but not so great for video.
Watching Videos:  :-( 
Reading:  :-(  
Writing:  :-)) !!


The Nexus 9 is not a perfect device. The aspect-ratio problem is something you can probably live with if you choose to buy the device and it's hardly going to be a surprise to you. The other issues are a bit less obvious when making the purchasing decision.

Lollipop is not ready

This summer Google announced Android 5.0, the biggest update to the operating system since its launch. But as any software engineer can tell you with big updates comes big bugs. The Android 5.0 Lollipop release is no different. 

The latest Android operating system version looks gorgeous. It is a giant leap forwards in terms of design and functionality of Android. Unfortunately the big change is still ongoing. Some things are still laggy and broken. Some design decisions of the OS version were ill informed and will be fixed later (or sooner, hopefully). Personally, I'm not too bothered with these issues as I have trust to Google that the situation will improve very fast. But you have to decide yourself if you're happy to tackle these issues or if you want to wait for the inevitable first iteration of updates to the Android 5.

Android 5.0 Lollipop is a typical .0 release. It has tons of problems that needs to be fixed but that is not uncommon in big software updates. The 5.0 shows us where the OS will go once this release is ready (probably around 5.1 or 5.2). I can forgive these issues. They will be fixed. Android 5 will be amazing once complete!

Lollipop now: :-(
Lollipop in future: :-) 


Trying to find the power button on a device like the Nexus 9 can be annoying. The button quality is also a bit iffy. To remedy the awkward interaction the Nexus 9 support double tapping on the screen to wake the device. While a good idea it is so unreliable that it's difficult to trust it. It works for about 75% of the time and the rest of the time I find myself first tapping the screen like an idiot and few seconds later cursing to myself and stumbling for the inconvenient power button. 

Double-tap-to-wake: :-( 

Build quality - USB connector

There are some unfortunate hardware problems with the device. As a Nexus device developers will be one of the main group of users wanting to have the Nexus 9. Unfortunately the USB port on the device is very flimsy. On a developer device this matters much more than on other devices as developers will be plugging the device in and disconnecting it all the time. Most Android devs can tell you stories of having to replace their devices after the USB connector wear preventing them from using the device as a developer device anymore.

USB connector: :-( 

Android issues - lag

Then there's something that has been plaguing Android since the very first versions and still is. Rotation lag.

When you rotate your Android device it will recreate the current activity and inflate all the views. It is a technical design decision that has made the OS very flexible but to users this shows as a delay whenever rotating the device. 

On a 200 euro device issues like this rotation lag can be ignored or at least accepted. The Nexus 9 is a 500 euro device and issues like this becomes unacceptable

There's also some other issues in the Android 5.0 causing slowness when switching between apps. The Nexus 9 should have plenty of memory and processing power to make app switching faster. Especially bad experience happens when the launcher app is thrown out from the memory and user presses the Android home button. In some occasions it can take more than a second for the launcher screen to appear and the home button pressed state to disappear. On a high end device like this that is something that I cannot accept. 

Both of these issues have the same root cause. Creating Android activities is slow and heavy process. Solving this is a difficult task but before there's a solution Android, especially Android tablets, will keep feeling laggy to users coming from iDevices. For the rotation issue we should be looking for some kind of workaround as it is so deeply based on the core architecture of the OS. Maybe we could use some temporary display giving the users feel of more immediate rotation.

For the home screen loading issue maybe another solution could be possible. I believe that the launcher app should receive priority treatment from the OS and maybe it should only be thrown out from the memory at the very last resort (which should never happen on a high end device like the Nexus 9)

Lag: :-( 

Then there's the speakers

Nexus 9 has two front facing speakers. After using it for a while I cannot understand why all tablets are not designed with front speakers. Watching some Netflix in the bed with my girlfriend is great. The sound quality is fine for the purpose and there's plenty of volume for 2 people watching the tablet same time. From now on front speakers will be one of the major requirements for me when buying a bigger tablet. iPad users, you don't know what you're missing.

Speakers: :-) 

Nexus accessory problem

Nexus line of devices might have unified style and software but accessories is something Google keeps continuingly failing with. The Nexus 9 keyboard case sets you back almost 130 euros. That is a lot for a keyboard. In the case of Nexus 9 it is a mandatory part of the device though and I believe that Google should not sell the device without it at all. But what about future? 

Except for few Nexus accessories they have all been compatible with just the single Nexus device they were built for. That's fine when the accessory is a case costing 20 - 30 euros but a for more expensive accessories users expect to get more. 

Compare this situation to iPad accessories. There are tons and tons of iPad accessories, both first party and third party. The reason is simple. People buy them. And people buy them because they can rely their accessories to be usable for the next generation device as well.

With Nexus your accessory selection is poor as 3rd party manufacturers are reluctant to get into the accessory business when the accessory will be viable purchase only for month or two (or a year in the best case). Google will remorselessly end-of-line their last year's devices and the new device will has nothing in common with the last one. All accessories will be obsolete.

I'm hoping to get enough life out of the 130 euro keyboard but I can't blame anyone thinking otherwise. Some kind of possibility of the keyboard being compatible with the next Nexus 9 (we don't even know if there will be one) would certainly help convince people to buy the keyboard.

Accessories: :-( 

Conclusion, final thoughts

Nexus 9 both is and isn't a great tablet. If you look at it as a 500 euro travel companion for writing it is awesome. If need a device for watching videos and reading on the go Nexus 7 beats it's larger cousin in pretty much every case. In my opinion buying the Nexus 9 without the keyboard case doesn't make any sense. Without it, it is uncomfortable to hold, unusable as a writing tool and beaten by the cheaper Nexus 7 in most use cases. Don't calculate the device's price separately. As I mentioned above, I don't think it should be sold separately from the keyboard at all... because what's the point. 

I love my Nexus 9. To me it is near perfect device for what I wanted it for. Your mileage might vary. I won't be recommending this device without reservation like I would the Nexus 7. But if you travel a lot and need a device to complement your heavy and large laptop and plan to use if for writing the Nexus 9 is probably the best device out there.. by far.

Nexus 9: :-)

Monday, 3 November 2014

Sensor Dashboard

- a developer tool for visually inspecting Android Wear sensor data

Gestural interfaces and activity detection is going to extremely interesting topics when we're starting to see the next generation of Google's Android Wear devices and apps that innovative developers write for them.

Over the last weekend after the awesome Droidcon UK conference there was an Android Wear hackathon. In the hackathon I got together with two friends and we put our heads together to build a tool for developers to help them understand what kind of data is available for them from Android Wear devices.

Here is what we built:

What does it do?

First, I'd like to remind you that the app was built in hackathon. It's not perfect. It is potentially buggy.

The app reads all the sensor data available from a connected Android Wear device and graphs it a connected handheld app.

The idea is to help developers to figure out if the app idea they have is feasible. Let's say that you're planning to build an app that detects push-ups. Without writing any code you simply install the Sensor Dashboard do pushups and observe the effect on various sensors. If the result of the test is that you see an identifiable peak of data on some of the sensors you can probably detect it programmatically as well and you have a good starting point for starting to design your algorithms.


This is how an airplane takeoff looks like in sensor data:

Future development

This app was built as part of a hackathon in just under 2 days. It's far from perfect. The UI doesn't really work very well. In fact, it's never been tested on anything else than Nexus 5 running Android 5.0 preview.

There's many features that we would like to see added to the app. Firstly, it would be great to be able to run the sensor data gathering also on a handheld device. This would not be too large change as Android Wear runs Android so the same service could easily be executed on a handheld as well.

I'd also love to implement feature to allow the user to pin multiple sensors and overlay them. I believe that in many use cases use of multiple sensors is required to make the gesture or activity detection easier.

Get the source

The project is open source and you can download the source code from github:

Please contribute back if you make improvements. Pull-requests are very welcome!

Get the app

The app is also available for free from Google Play store:

Please give us feedback!

I believe that this app can be pretty useful tool for devs. It would be great to hear what you think about it and what could be better. Feel free to ping me at Google+ for any questions: 

The team behind this project:

Monday, 8 September 2014

Connected Bicycling - Part 1 - Runtastic Bike Sensor

Since the launch of smartphones (or at least the modern smartphones) one of the largest categories of apps and accessories have been related to fitness, usually running or biking.

Download numbers of these apps and their free versions are in millions. People use these app. Smartphone focused fitness has become so big business that all these competing companies are looking into figuring out how to increase their revenue and how to lock users into their system.

A good example if Runtastic who have created an ecosystem of connected devices around their smartphone software. They sell everything from bathroom scales (which is still not supported by Android for some unfathomable reason...), to biking cadence sensors and new wrist bands.

As a smartphone geek and cyclist this combination fascinated me. So much so that I activated a Gold account on Runstasic and bought one of their Bike sensors to attach to my bike.
My geekiness, and enthusiasm about connected biking is what drove me to write this blog post. In this post I want to shed light on some of my experiences with the current state of the Runtastic biking sensor as well as explain some thoughts about improving it. 
Before I go into any more details about my experiences let me explain few things about me. Different people have different needs.

I am a hobbyist, free time bicyclist riding a cheap (but good) mountain bike. I don't care about training to ride, getting the best results or the fastest times. I never compete in any biking events. I ride my bike accompanied by my girlfriend to enjoy scenery and visit new places around the Bavarian Alps. I am a Sunday cyclist.

My setup

My current biking setup is simple. I use a water resistant container to hold a cheap Android smartphone (Moto E) as my main screen. In the same bag I have a battery bag that allows me easily go a day with the phone screen on and on full brightness. On the Moto E I run Runtastic Mountain Bike PRO app to display my route, speed and other information. The Moto E is connected to the Runtastic Speed and Cadence Sensor via Bluetooth.

I will be extending my connected biking by adding a Lifebeam Helmet once it arrives.

Runtastic Speed and Cadence Sensor

As I mentioned, me being an Android geek I wasn't able to resist getting my bike connected. I bought the Runtastic Speed and Cadence Sensor.
To me this was a no-brainer. Even though I don't care about training or improving my biking I want the data. I don't know what to do with the data but I want to have it to figure out what to do with it later.

You can buy the sensor from (but read below first):

Unfortunately, the sensor isn't what I wanted it to be. It's a painful reminder about two things:
  1. Building hardware is difficult.
  2. Bluetooth is very unreliable and difficult to get right.


The sensor part is all in one section of the hardware. The cadence sensor is larger and bulkier and attached to it there is an adjustable arm for the speed sensor. Both sensors have counterparts that are attached to the pedal and to a spoke.

When attaching the sensor you're going to run to the first problems of the system. There are thousands of different bike frames and Runtastic has tried to create a sensor that can be attached to any of them. The sensor can be attached with a fast attach rubber strap in the package or cable ties (also included). On my bike it was literally impossible to use the fast attach strap which, in fact, broke immediately when trying to attach the device. I ended up attaching it permanently with cable straps.

The sensor and the counterparts have to be very close to each other for the system to work. The manual says that the distance has to be at maximum 4-5mm. That is very close so you have to be very precise.

Setting up and connectivity

The next step after setting up and performing the initial connection to the Android app it was time to for the first ride. The amount of frustration during the first ride cannot be oversted. The sensor seemed to be connected but might not work. There's no indication on the Android app if the speed sensor works or not (you get GPS based speed information if the sensor fails).

The feeling is probably best described as helplessness. There was nothing I can do. I tried to reconnect, rebooting phone, restarting everything and even removing the battery from the sensor. There was no way for me to tell if the sensor wasn't on (it has no indicator for it and it is supposed to turn on automatically), if the Bluetooth connection failed, if the sensor wasn't correctly setup to read the counterparts.

A lot of this frustration could have been alleviated with a clear user interface on the smartphone app telling me what is going on!

After fighting with the sensor and rebooting everything multiple times it finally woke up. I still have no idea what actually caused it to come into life.

Here is a an example dataset from a short test ride with the sensor connected.

From the chart you can see the value going up and back to zero. Many of the zero points are accurate but I noticed the sensor losing connection many, many times which caused the cadence value to drop to 0 even when pedaling. Also worth noting that the last 10 minutes of cadence data is completely missing.

So much for reliability. 

Few weeks later for a longer weekend ride things were even worse. The device stopped working altogether. It simply refuses to connect anymore. After realising that the system wasn't working I decided to ignore it as I didn't want to spoil my good mood by spending hours fighting with the device. So there. The device has now became a thing that is attached to my bike but doesn't do anything. So much for gathering any data.

As a sidenote I want to point out that my phone's Bluetooth works just fine. I have an Android Wear device connected to it all the time.

Conclusion about the Runtastic sensor

So, what to think about this device? It clearly isn't very good. It is difficult to install, unreliable to use and frustrating. It also isn't cheap. So why am I not rating it with one star on Amazon and moving on?

I like the device.. well, I don't like the device but I like the idea of the device so much that I want to see it work. I want to see a second version of this device with improvements. I will even spend the time and frustration to get the damn thing working again!

How I would improve the sensor?

I think the sensor has two massive problems. The first is its lack of feedback. There's no way for me to know if it is out of battery or if it'll soon run out of battery. I cannot know if it is on or if it is connected but not working. This needs to be absolutely fixed from usability point of view. The frustration must be eliminated.

Another big problem is the massive data loss with the device. It wholly relies on the smartphone to keep the information and the device itself doesn't have any smart component. This leads into a situation where if the connection to the smartphone breaks all data for the duration of the connection loss is lost. And lets face it. Bluetooth sucks. There is no such thing as reliable bluetooth connection. This must be fixed as well for the device to be pleasant to use.

Fixing these issues

I think Runtastic has to give up the idea of keeping this sensor so simple. It's too simple to work.

I would like to see it separated to a smart component and rest of the sensors. I'd maybe even separate the two sensors to a separate pieces of hardware to make it better compatible with rough terrain biking.

By the smart component I mean a piece of hardware that is physically connected (either directly or with wires) to the sensor components. I want the smart component be powered with a rechargeable battery as well as contain at least one indicator LED. The LED should indicate status of connectivity as well as battery status.

I want the smart component also be detachable (for recharging and anti-theft).

It need to have small amount of memory as well so it can keep all the data when there's no connection to the smartphone. This way I will be able to get my the information to my phone later even when I have issues with the phone connection. It should automatically sync between the phone and the sensor whenever connection is reconnected.

The smartphone software needs to be improved as well. I want to have clear indication about the status of my sensor. I want to know if the speed I see is from GPS or my speed sensor and if my cadence sensor is connected or if the sensor is connected but reporting 0 cadence.

In short I want the sensor to:

  • Better indicate if it is on, connected or out of battery.
  • Recover from Bluetooth connectivity issues.
  • Have clear feedback on my smartphone display of the status of the sensor.


The Runtastic sensor is not very good but the idea is. I have never before been such a fan of a device that clearly doesn't work. I feel that this is the beginning. I want to see where connected bicycling (and health) can be taken.

I will be writing a follow up to this post about my thoughts about using the data and what I'd like to see in the services for hobbyists.

Stay tuned.

Saturday, 25 May 2013

Take two on OUYA

I wrote some first impressions of the OUYA console week or so ago. You can find those thought here in the previous post. I promised to write more impressions once I get to test it properly. So here goes.

OUYA, the party console

After a short while with OUYA it becomes painfully clear that the console is not very good single player experience. The gaming offering is full of not-very-good retro titles and simple casual games with very derivative game-play ideas. I have no doubt that some of the negative reviews and posts about OUYA are because the people testing the console stopped here. You can't say that they're wrong but I think they've missed the point of the console. 

If single player experience sucks it doesn't mean that the console sucks. It is actually awesome. The key is multi-player casual gaming. 

OUYA game offering pre-launch

Consoles live or die with their game offering. Nobody will buy a console without games. At this point it is clear that OUYA is falling short from the estimates of number of launch titles (400).

I've always found the number of titles irrelevant. To me (and I assume to most people) what matters is the quality of the titles available. I'll take one good title over 1000 mediocre titles. It looks to me that OUYA might have exactly that.


To me, BombSquad is the killer title for OUYA right now. It is extremely polished, visually pleasing party game. It is also only available for OUYA and OSX making it exclusive for OUYA in the Android world.

The player allows 1-8 players to fight it out by throwing bombs at each other. The varying minigames keep the game interesting and interesting power-ups add more to each level. 

As an added bonus you can also use your Android phone to control the game in case you lack the sufficient number of controllers with the BombSquad remote app

Ice Rage

Ice rage is an arcade 1v1 ice hockey. The gameplay is extremely simple but enjoyable. While this game is definitely something you won't play whole night it's definitely an entertaining title in short sessions. This game is also available from the Play Store although I suspect that it isn't a very good tablet / phone game. It's definitely a multi-player focused title.

No Breaks Valet

A casual game doesn't have to have mind-blowing graphics to be great fun. No Brakes Valet definitely fits this category. It is crazy fun as a two player game.

Buying games from OUYA store

In the previous post I said that I was not planning to buy games from the OUYA store and was going to stick with the Play Store. Well.. I've changed my mind. I already bought couple of games.

So, what made me change my mind? It's simply because the games I want to play on OUYA are very different form the games I want to play on my other Android devices. My initial objection for buying from OUYA store was that I didn't want to buy games twice. But that's not the case. For OUYA I buy multi-player casual games for my other devices I buy games I can play alone.

Games for OUYA also seem to have much more reasonable approach to the payments. Recently, the Google Play Store has been plagued with the idiotic pay-to-cheat or cash-cheat method of payments. I simply refuse to pay that way. On OUYA, the games actually get you more game play when play. You  buy games you want to play unlike in the pay-to-cheat model where you pay to complete the game faster. I think that currently the ridiculousness of the cash-cheat model on Google Play is really helping the chances of OUYA being successful.

OUYA == Multi-player-only

I can't emphasise this point enough. OUYA is local multi-player only at least for now. In fact, so much so that OUYA is shooting themselves in their own foot by selling the console with only one controller. They should not let anyone experience the console without out-of-the-box multi-player experience. It is a party console.

Game discoverability

90% of the games in the OUYA store are not very good but as we've seen there are good games for the console and we'll probably see even more good titles before the console's consumer launch in June. But how can I find them. As I mentioned in the previous post the categories in the OUYA store are confusing. Finding good titles is currently nearly impossible.

I really want to see a more carefully curated store. I think this is the key for OUYA's success. If they manage to get more great games to the store the users must find them within the first few minutes of use. OUYA should really add a "Best multiplayer" category to the main level of their store.

Another thing that they need to add before the launch is multi-player information to the game info screen. We need to have a new bullet point in the Overview telling the user how many players the game supports. In the case of BombSquad it should read something like "for 1-8 players".

Some Version 1.0 issues

The pre-release console suffers from some critical version 1.0 issues which OUYA must fix before the release.

OUYA controller

Firstly, the controller analog stick quality. Without better control many, many games will not feel great. I believe that it is impossible to create great racing games, for example, right now until the controller is fixed. One game greatly suffering is Flashout 3D. Controls in the OUYA game are very bad and the game experience suffers greatly. The same game on an Android tablet on the other hand is very smooth and the controls feel fine. To me this is a red flag that the OUYA controller is not ready for release.

Fortunately, according to this OUYA forum post it looks like this issue is currently getting the attention it needs and is likely going to be fixed before the consumer launch.

3rd party controller support

I'd be willing to bet that many households getting the OUYA already have either a Playstation or an XBox. As many people will be buying the OUYA with only one controller it is very important for OUYA to make sure that people can use their existing controllers with the system. I have a PS3 and I managed to pair the PS3 controller with my OUYA without big issues. I'm a geek though and to me it was no issue to jump into the settings and pair the controller manually via the Android's bluetooth system and to plug it into the OUYA USB port as hinted on some forums. But I believe that for many people this is a step too far. The OUYA really needs to put a simple instruction screen in place for pairing common 3rd party controllers with the box. I believe that this would make the console much more appealing and allow more players to experience the multi-player experience.

OUYA should also make sure that the most popular controllers do work with the games they approve to the store.

Random controller disconnects

This could be another pre-launch hardware issue but a greatly annoying one. The OUYA controller randomly disconnects from the console. In a two hour gaming session the controller disconnects maybe ten times. This is too much and disrupts the experience. OUYA needs to fix this to never allow the connection to be disconnected.

TV overscan

If you have ever built any software for Google TV you're very aware of TV overscan issues. Basically this means that some TVs don't actually draw all the pixel they promise to draw. There's a long explanation of the issue in the Google TV developer site where you can get more information.

This means that apps cannot use the whole display and should leave the very edges empty. Currently, this isn't happening though and most games are still built for tablets where drawing all the way to the edge is OK. Effectively this means that at least on my TV most games have part of their title or controls cut of. It makes the games feel less polished. I doubt that there's anything OUYA can do on the software side but this is something that must be pointed out to all developers wanting to put their games on the store.

Conclusion and TL;DR

I was positively surprised by the console when I first tried it and once I got to try it with my friends in a party I was convinced that the console can be great. The killer games are there but because they are casual games one or two killer titles won't make it. There have to be many more. For this OUYA must make tweaks to the game discoverability on their store. I'm 100% convinced that this is a platform that can succeed. There's only few issues that must be sorted out and the console will feel like a great product. It is so close that the goal is clearly visible now OUYA just has to take the correct last steps to make it. I'm, for one, hoping that they will succeed.

Monday, 6 May 2013

OUYA Awesome, Mediocre, Sucks and Everything in Between

OUYA is one of the all time Kickstarter succes stories. Their campaign got funded in matter of hours and ended up going all the way past $8.5 millions (from the initial goal of $950.000). That meant that the company was able to change focus from creating a production prototype to creating a production run.

Of course the folks at OUYA didn't rush to create a final product first but created few iterations of dev devices before sending out the backer devices. Even the backer devices might not be exactly the same as the first devices that are sold directly to consumers. This is the only way to go. Hardware projects will always have issues. That's also something that anyone backing a startup to create a new device will have to know and expect.

We've seen people writing reviews of the OUYA sent to backers and even giving them rating as if it was a final product. That is ridiculous. Any site giving a number score to a beta device is doing it wrong.

That said this article is not a review. This is just my thoughts of the device as well as my thoughts about backing hardware startups through Kickstarter.

Kickstarter experience

When I heard about OUYA it took me about 5 minutes to decide to back it. I did it without thought. I wanted this device to exist so I put my money down. That is what Kickstarter is for. It is not a pre-order system for things that are going to be created anyways.

There's always a risk in this kind of transaction. I was paying for something that didn't exist yet. I gave the company money for their promises.

As anyone know talk is cheap. What am I entitled to expect? If you've ever worked in a software project you know that the things that seem to be simple and plans that were set to stone at start of a project will change. With hardware I imagine that it happens even more.

I think that the general cynicism soon after the OUYA Kickstarter made big headlines reflected the fact that promises are difficult to follow. Some went even so far as saying that the OUYA Kickstarter was a scam and we'll never see the box. I'm happy that the most cynical people were wrong and the console is now getting shipped to backers.

Backing a US company from Europe

This is a topic I have to bring up even though this isn't necessary something you can hang on OUYA's neck. I live in Europe, Germany. This causes a lot of issues. If you're an Android users you know how often Europeans feel like second class citizens of the Google's Android ecosystem. But it's not only that. Ordering anything outside Europe is likely to cause issues. You need taxes, adapters and high postage.

What did the OUYA actually cost to me?

I paid $99 for the actual console. On top of that I had to add $20 for postage to Germany. When it arrived here I had to pay import tax of 19% which is slightly more than $20. I also had to go an pick it up form the customs office outside town costing me two hours of lost work and public transit costs but let's not count them here. Unfortunately, the expenses didn't end there. The US power plug that comes with the console is not usable here without an adapter. That is an extra $10. So my $99 console ended up costing $150 instead. 50% extra over what someone in the US would have paid to back OUYA. Is it worth it?

The answer is no. Because OUYA got backed anyways my backing didn't make any difference. I would have been much better off waiting and getting an European version once available. Spending extra 50% to back a startup does not make any sense.

I'm afraid that this was the last hardware Kickstarter that I'll be backing. This simply isn't worth it.

As I said this is hardly OUYA's fault. They could have, however, made it easier for us. They could have imported it in Europe (but probably too time consuming and expensive) but at least they could have included an European power plug (but maybe this would have had some legal implications).

[UPDATE - 11. May] I've been informed that even the backer edition ships with local adapters. Me getting a US one is an exception.

Hardware quality

OUYA is cheap. It is a $100 device. Comparing it to more expensive consoles doesn't make much sense. So let's just look at it as a $100 device and see how it feels.

The box

One thing that is immediately clear when you take the OUYA out of the box is the nice feel of the actual OUYA box. It feels durable, solid and looks nice. I have no problems putting it in my living room next to my TV. It feels so rugged that I'd expect it to be a good travel console that should easily survive a trip in a backbag or suitcase.

The controller

The box doesn't really matter to me. I don't hold it and if it were ugly I could hide it behind my stereo set. What matters is the controller.

A screen capture from the OUYA Kickstarter video.

It looks to me that the OUYA team already had a very complete idea when they introduced the project. The completed controller looks very much like the ones shown in the Kickstarter video.

It looks like a fairly standard console controller with the two analog sticks, bumper triggers, d-pad and few buttons. Unfortunately, the controller doesn't have the same quality feel as the box itself. This makes me wish that more time would have been spend with the controller and the box would have been left more simple. The controller is what I'm holding in my hand all the time. There's a feel that the parts don't quite fit together. But it is no disaster in any sense of the world. It is something that can be expected from a $100 device. As a backer I'm definitely fine with the feel of the hardware. It just feels like the controller was made by a different company than the box.


Unfortunately, there's an issue with the controller. The D-Pad is basically broken. It reacts to events very randomly and is nearly unusable. This nowhere more apparent as when trying to type with the onscreen keyboard. There is no difference in feel between successful and unsuccessful D-Pad press.

This is very unfortunate. In the first few minutes of use of the console you have to type in wifi password, account name and account password (or even worse register a new one). D-Pad is the primary control for the keyboard. The analog sticks aren't accurate enough. This raises the rage-meter of any user and gives the console an extremely bad first impression.

Analog sticks and buttons

The analog sticks on the other hand work nicely. They reach to movement as you'd expect. I do suspect that these aren't real analog sticks though. All games I tries only have few directions instead of infinite like on other platforms. I'm not sure if this is due to the hardware or if the games don't have yet implemented proper controls. 

The buttons on the controller, including the bumpers, feel very nice and have a good feedback.


OK, let's dive into the software then. As established the hardware is something that can definitely make this device a decent experience. But how does the software stack up?

But before we go into details please keep in mind that the console I have is still running pre-release software. The OUYA team is still working on making final tweaks and killing bugs in the system. That's why in this post I'll ignore some random crashes and issues I had as I have faith in the OUYA team for these issues to be gone once the consumer version launches.

OUYA home and OUYA Android skin

OUYA has replaced the Android home screens with their own custom launcher as well as lightly skinned some of the components to match the OUYA theme. Skins are usually bad (think of Sense and Touchwiz) but not so in this case. Android has not been designed to be used on TVs and with D-Pad being the primary navigation. Running stock Android would be a bad idea.

The OUYA UI design is definitely influenced by other consoles and that's good. There's only so many ways you can build a UI for D-Pad. The OUYA UI is good. In fact, it is very good! It responds to your commands very well and you'll never get lost in the layouts. It is visually pleasing with a very graphics heavy style that works perfectly on an entertainment device.

There are some small issues though. Firstly, I had no idea what these categories were. Is "Check it" like featured? What about "Favs"? I only know what the "Sandbox" is because I watched an OUYA interview few months ago where it was mentioned. Maybe they're trying to be a bit too creative here.

Another issue I have is exiting games. The controller has an OUYA button and it takes you out of any game if you long press it. Unfortunately, it only works after you release the button. This leads you pressing and holding it and thinking if it might have been long enough and releasing to check and repeating until it works. This should definitely react immediately one the button is held down long enough.

Almost a full skin

You still endup in unskinned Android screens now and then and especially see some stock Android popups. But I have a feeling that these will be disappearing one-by-one as soon as the OUYA team gets around to implement them.

I thought I'd never say this but I much prefer the OUYA skin and wish it would cover everything on this device.


The console is not limited to only games. You can find apps from the OUYA store as well (at the time of writing this there are 3).

I was exited to see in the OUYA store. It was also mentioned in the Kickstarter video. It's not good though. has decided to give OUYA their phone app. It is very near unusable with D-Pad. I know because I've side loaded this app to my Google TV. There's a good reason why twitch has not made this available for other TV platforms. The app simply sucks on TV. I really hope that Twitch is working on a proper app and this is only a placeholder until the real app is out.

The other apps aren't much better. tune in has a radio app. This is, in fact, the only music player available for OUYA at the time of writing this. The tune in app is a tablet app. Not designed for D-Pad either. This is pretty much unusable as well. I know that you can use the OUYA's touch pad feature as a mouse but you shouldn't have to. It's a very bad experience that should only be a backup.

So the OUYA apps are bad. They should be designed for TVs. Google has been trying to crack this issue for years now. OUYA should point their app devs to the Google TV design guidelines.


But what about games then? They are the core of OUYA. OUYA will either live or fall with its games.  At the time of writing this I've only tested about 10-15 games so I will not be making any statements about the general state of the OUYA game offering. I'll do that in a separate post once I've had time to really dive in and find games I like.

I'd like to explain some observations though.

Analog sticks or arrow keys?

I already mentioned this when I talked about the analog sticks. Physically they feel nice but in-game there's something wrong in all the games I've tried. They don't actually seem to be analog sticks but 8 direction arrow keys instead. Playing a game with the analog sticks feels close to PC gaming with arrow keys.

This becomes very apparent when playing games where you control a space ship, tank or whatever from a top-down perspective. While on other consoles you've got used to a nearly infinite directions you can choose. 

I really hope that this isn't a hardware issue and the games simply have not implemented this correctly.

The Secret of Universe Alpha


Slow response bumpers

Another worrying thing was a laggy response in this pinball game. I didn't notice any lag in other games so this might just be a special case. There's minute but noticeable delay between me pressing the bumpers on my controller and the pinball game reacting. This could also be that the button travel on the bumpers is pretty long. It could be just that you need to get used to it or maybe the game just isn't well built. The game felt fast everywhere else which lead me thinking that there might be something wrong with the bumper buttons.

Pinball Arcade

Need for guidelines

The games are in a dire need of guidelines. This is a general issue of Android platform where all games are very unregulated and every developer seems to build their own menu structure and reinvent the way game is exited.

I wonder if OUYA could try to create some standards to the game UI. I'd love to have consistency in the menu navigation, settings, pausing a game and exiting it. This is much easier said than done though.

Vector contains a ghost of Gingerbread

Some games I liked

Some tech sites in their early reviews kept telling that there's no good games on OUYA. It looks to me that they were looking for Call of Duties or Halos and left out everything less than AAA. I definitely found some casual games that were entertaining. None of them were perfect but to me these titles prove that OUYA is very much capable of running games with great gameplay!

Here are some of them.


Ice Rage
Natural Soccer

No Play Store

OK, now the elephant in the room. There's no play store.

Vendor lock is evil. But everyone is doing it. When I decided to start using Android phones I got locked in to the Google's Android ecosystem. Any software I buy from the Google Play will always be available only on my Android devices. But this vendor lockin isn't as bad as some others. I'm locked into a platform not to one manufacturer. I would never buy any apps from Samsung's or HTC's store to use on my phone. However, I've bought games and apps for probably few hundreds of euros so far from the Google Store. The reason I'm fine with Google Play Store is the promise of these apps working in the future on any Android device (with Google Play) I decide to buy.

Unfortunately, OUYA is different. Their store is OUYA only. If you buy something there that is the only platform you'll be playing the game on. Making matters even worse many of you reading this have probably already bought many or at least some of the games available from the OUYA store. And no, you can't play the games on your new Android console.

I do understand that it is up to Google to give devices certificate to run Play Store (and ship with the rest of the Google services). The current certificate program probably doesn't allow devices with OUYA's approach to get the certificate. So this is another issue that cannot be fully blamed on OUYA. Still, to me this is one of the issues that might limit my use of the device and most likely cut it very short. I won't be buying games from the OUYA store. At least unless they're done especially for the OUYA controller.

OUYA vs. Google TV

Comparing OUYA to XBox or PS would not be very beneficial. Not because OUYA is cheaper but because OUYA is clearly targeted more towards casual gaming. Comparison to Nintendo Wii is something that will probably make more sense. OUYA is in the same price class as Nintendo Wii (both about 150 euros). But then comparing OUYA games to Nintendo Wii games would not make much sense as games for one cost up to 60 euros and for the other only couple of euros.

That's why I'd rather draw a comparison to Google TV. The latest generation Sony Google TV box currently costs about 150 euros here in Germany. So the price comparison is fair. Google TV also has games, same games in fact, which means that the games are priced very similarly.

Google TV software vs. OUYA software

At the time of writing this Google TV still runs two year old Android version (3.2) while OUYA is on almost the latest (4.1). The same story continues throughout the UIs. Google TV is difficult to use, old UI paradigms (it still uses a menu button) and ugly visuals where OUYA's UI is simple, beautiful and easy to use.

OUYA home screen 

Google TV home screen

OUYA installed apps screen

Google TV installed apps screen
There's really no question about it. Google TV UI can't hold a candle to OUYA's.

Then there's the issue of controllers. One of the controllers in the picture below has been designed its users in mind and the other has been designed without knowing what it is for. OUYA has a lases focus to be an entertainment device. Google TV tries to be everything. I have always said that if your TV remote has a "Ctrl" key something is wrong. I recognise that there's a market for Google TV and that's probably exactly the same market that want to have a living room PC. Most people don't. I like devices that have focus. You usually endup with much better design that way.

Gaming on Google TV vs. OUYA 

This a no-contest really. There's still no decent Google TV hardware out there. Every single device is underpowered and struggle running heavier apps. Some games do run on Google TV but most of them don't. Many developers have not bothered to bing their games to Google TV because there's no point really.

Apps on Google TV vs. OUYA

There's not many apps on OUYA right now but it would be unfair to held it against it before the launch. The situation might get much better in the coming months. The apps OUYA has now is giving some cause for worry though. They are not TV apps. On Google TV you find apps that are designed for TV. There aren't many but there are few. OUYA should really be talking to those companies who have created a TV friendly UI and ask them to put it available to the OUYA store as well.

But one big is missing on OUYA and probably will be for a while. Google. I need to have YouTube and Google Music on my TV device. These are the apps I use in my living room and can't live without. If they were on OUYA there would be no question which is better Google TV of OUYA. But before that I think the question will remain unanswered.

Google TV YouTube app. OUYA needs a YouTube app as well!

Google TVs YouTube app allows seamless communication from phones and tablets. I'm not sure if I can live without this anymore.

Final thoughts

I have not used OUYA nearly enough to really review it or present any conclusions about it. Don't base any purchase decisions to this article. This is early thoughts and observations. I will be following up this article if I find time to write again with more in-depth review.

OUYA has definitely been a positive experience for the short time I've had it. It's far from perfect and there are some big problems but the start is promising as well as impressive. A startup creating something like with crowd funding is a feat. I'm happy that I have been a tiny part of it.

I find OUYA important and awesome in one more perspective. It proves that Android truly is Open Source even if Android ecosystem isn't. I'm pretty sure that one of the very first tasks the OUYA tech team did was to follow Andy Rubin's definition of open. I leave you with that. Good job OUYA team!